laterstars blog

why laterstars is what it is - the laterstars manifesto, pt. 3

This is the last entry in a series about why laterstars came into being. Part 1 sets the stage, going over all the problems I ran into while trying manage tons of information from hundreds of RSS feeds, and Part 2 talks about why I think there needs to be a shift away from RSS altogether. In this final post, I’ll explain how laterstars came to be, what I’m trying to accomplish with it, and why I think it’s the right solution for me.

The Vision™

Ok, so what’s the one thing that pretty much every Twitter client under the sun supports besides the general tweets and replies and stuff, but is often overlooked? Favorites. There’s built in “integration” for faves in the web interface, mobile web interface, Tweetie, Twitteriffic, Echofon, TweetDeck, Seesmic, and on and on and on. But wait… favorites? Aren’t those for marking tweets you think are funny from Merlin Mann and all the other aspiring Twitter-based standup comedians to let them know that Twitter likes them, it really really likes them? Eff that. I mean, sure, I’ll fave a witty tweet here and there if it’s exceptional, but otherwise, what’s the point? To put it more bluntly, what’s in it for me?

Historically, I always found Twitter faves to be a little bit pointless, but harmless… one of those features that probably made it in because it was relatively easy to implement, and, well, why not? So I pretty much ignored it for the last 3 years or however long I’ve been on Twitter. Like I said, though, it was definitely embraced by the comedian set, and taken to prominence by the recently defunct Favrd and more recently sites like Favstar and Tweeteorites, where the tweets with the most faves get prominently displayed. To me it seemed like a huge group hug, a big virtual clique that turned into a kind of weird back-patting/support group session where people circularly faved each others’ stuff even if it wasn’t all that great, and you had better fave everything the cool kids tweet or else. Granted, I’m probably just jealous because I’m not funny enough to be part of the gang, but even Dean Allen, the pioneer of these types of these sites, had some choice words for what his creation had become:

… Sites like this one now serve mainly as fuel for emotional up-fuckedness in the guise of a game.

Just an idea: next time you see something you like, write the person who made it a note telling them so. Even better, explain why.

Now, personally I don’t think anything is necessarily wrong with those types of communities… everyone needs to laugh, everyone needs validation, and I enjoy going through these sites and using the Favit iPhone app when I want to read some funny stuff. I’m just not ever going to be part of it and I’m not really interested in just piling on more praise to the people on the leaderboards. Does this mean twitter faves are useless for me?

No, of course not… the true genius behind faves is that they’re not actually mandated by Twitter to be for “marking tweets you think are funny so other apps can aggregate them,” they’re just kind of there. Why can’t we leverage favorites for something, well, more useful? I was definitely not the first to think of it, but the other usage of faves could be as a bookmark. I noticed that when looking at a lot of my friends’ favorites, most of them were usually tweets with links in them, not the pithy witticisms I was expecting and was used to seeing from the fave-oriented sites. I then checked out the faves page of one of the engineers who actually works at Twitter… lo and behold, it was mostly tweets with links. It seemed as though this usage was Faves For The Rest Of Us… not trying to boost anyone’s popularity or attempts at provoking someone to fave your stuff in return… just something quietly useful.

I tried it out myself, but my faves ended up in the same abyss that my GReader faves ended up. I just never went back to them. Why not? Well, first off, they were easy to forget about since the faves feature is kind of a second-string feature on the Twitter site and in clients. Secondly, since almost every link that shows up on Twitter has been shortened, you don’t have a good idea of where you’re actually headed. Some clients auto-expand them, and some even embed them if they’re images or videos, but for the most part you’re looking at a bit.ly url. And, finally, navigating and managing faves just isn’t that pleasant. I want a power tool for managing all this stuff, and a second-class feature just isn’t going to get that kind of attention.

But I knew there was something there. A lot of people were already using faves as bookmarks and just slogging through all these shortcomings. Some people got clever and subscribed to the RSS feed of their faves in GReader, but I didn’t really see the point in that… it would just be another feed and the posts were all just the tweets with their short urls. I tweeted in frustration that Twitter needed to provide some way to use favorites to bookmark links, and got some responses form other folks who felt the same way. One of those responses came from my friend Phil who said “this could be the project we’ve always talked about doing together.” And so it was. Phil was packing up his whole life then to move out to SF to work for Engine Yard at that time, so he couldn’t get started right away, but I was off to the races.

The Stuff That’s Important To Me

So I started working on the app right away. First things first, what to call it? Well, we were starring tweets to read the links later, laterstars.com was available. Done.

With that out of the way, what were the main goals going to be, what was the point? Seeing that I’m basically building this for myself, here are the things I care about:

• Links From Faved Tweets — duh. But, it’s good to set the baseline. The app is for extracting links from faved tweets and making them easy to go through and manage later. That’s it and that’s all.
• Keyboard Navigation — ok, so I’m a nerd, but dammit I loves me some keyboard nav. One of the main reasons i think GMail is great is because of the keyboard nav. The ability to get through your inbox super fast, deleting, labeling, archiving, all while not having to use your mouse, is a huge time saver. I wanted the same power as I went through my saved links. This is one of the things I think Google Reader actually got right, and it’s pretty much the only tool out of all the older options that has it.
• Powerful Management — on delicious, Instapaper, Pinboard, et al, if you want to delete a whole bunch of stuff, or mark certain ones read or archive them or whatever, you’re clicking on a whole bunch of individual links. Instapaper lets you archive all links older than 30 days or all unread links, but that’s not what I want. Want to delete 20 links you don’t care about anymore? Click delete, click yes it’s ok. Now do it 19 more times. Might not sound like a big deal, but to me it’s a deal breaker.
• Workflow Modeled After GMail — with these last two points in mind, it’s no surprise that GMail has been a big influence on laterstars since it excels at both. Besides those points, though, the whole idea of having an inbox where everything lands and then giving you the option to judiciously organize it by deleting stuff, archiving it, labeling it, etc, or, just leave everything right there, is part of the genius of GMail. You have both options, and both work just fine within their system. Some people would approach something like laterstars as an inbox that they need to constantly process and keep tidy (like email “inbox zero” or Instapaper), while others would just treat it as one huge receptacle of links they want to save (like delicious or Pinboard). I want both to be workable.
• Focus On The Links — short urls are really convenient and arguably are what make Twitter a viable link distribution platform in the first place, but when your whole context is a somewhat cryptic tweet and a blind url, it’s not the optimal situation. As an example, there’s this dude I work with named Cliff. Cool guy, but man, he tweets some rude stuff. Funny stuff, yes, but rude. Being able to fave his tweets, which are usually just a bit.ly link and maybe some vague explanation of what you’re about to see, to come back to later when I’m not at work would be a definite plus. That’s just one situation why I wanted the links’ page titles to be the main focus, with the full urls and tweets they came from provided as context. And a small preview would help too…
• Inlined Content — much of the reason i didn’t want to use Instapaper for everything was because I tend to want to save a lot of random images and videos that people tweet. With Instapaper’s focus on text, it makes it killer for reading articles, especially on the go, but not great for this type of stuff. Sure you can just click the links to open them in your browser, but I’d much prefer if they were inlined and available to view super fast right there in the list.
• Clean but Feature Rich Interface — I know some people prefer ultra minimal interfaces with just plain lists of text and nothing else. I see the appeal of that, but whenever I try using apps that attempt to adhere to that philosophy I always find myself wanting more… more functionality, more visual hierarchy and differentiation, more sense of aesthetics. I really enjoy interfaces that have a lot of thought put into the details and usability while still being able to balance it with styling that evokes a sense of beauty. Products from Apple, sites like Tumblr, Dribbble, Gowalla and a bunch of others, and much of the OS X indie software community fit that type of aesthetic pretty well. Products like Google and all their apps, Friendfeed, Pinboard, Instapaper, etc. take the other approach. This is all, of course, purely personal preference, but I’m building this for myself and I want a richer interface and I want lots of tasteful features.
• Twitter Only — I’ve explained why I think Twitter is the best source of links… having the app enforce this is a must to keep me in line and not let things get out of control. Freedom Through Constraints and all that.

And so far I’ve worked toward bringing all these points to life. Extracting links from faved tweets? Check. Keyboard nav? Check — just tap ‘?’ on your keyboard to see what’s available. Ability to do mass management of links? Check — use the keyboard nav (x, like GMail) or shift+click to select link rows and most operations can be applied to all of the selected items. Focus on links? Check. Inlined content? Partial check (still have more sites I’d like to include, but we have pretty good support so far). Clean, feature rich interface? Well, I think so, and the response has been way more positive than not, but this is one of those things where I’ll never make everyone happy, and I’m ok with that. And finally, Twitter only? Check. In fact, I recently had the temptation to veer outside of this belief with the idea of importing stuff from delicious/Pinboard, but I went with the Twitter based bookmarklet instead. That’s going to make some people unhappy and it’s still a bit of an experiment, but that’s the way it’s gotta be for now. I need to go whole hog with this approach to see if it’s the right thing. As they say, Lean Into It™.

There’s still a lot of work ahead to make laterstars the silver bullet for all the issues I was having with the other apps and make it my One True System. For one, I’m still using Instapaper for reading articles on iPhone, and will probably make it easy to send links to it from within laterstars itself at some point. I am experimenting with article previews that auto-apply readabiity, that’s getting part of the way there toward nice text-only articles, but for now will continue use IP for what I find it good at. Do we have an iPhone-optimized version? Not yet. How about full text search? No, not at this time, but a friend just contacted me about some ideas he has for it. Rich social features? Well, you can browse other peoples’ links, and we’re working on being able to subscribe to them, but there’s a ton more that can be done with this. We have a lot of ideas, but it just takes time.

So no, it’s not perfect, but you know what? It’s working pretty damn well for me so far. I haven’t opened Google Reader since I started the project. When I’m out and I pull up Tweetie on my phone to check out what’s going on and I see some links that look interesting, I just swipe and star. Super fast. Same thing when I’m using Echofon on my Mac, just hit ‘F’ to star and continue on with my business. When I get a link in IRC that I want to save, I use the bookmarklet to see who else has tweeted it and fave it right there. When I get a chance, I open up my inbox and start quickly going through links. I even browse through other peoples’ pages a lot too, and have found some pretty awesome stuff I would have never seen otherwise. I’ve also seen tweets from some other folks who’ve said they’re pretty close to getting rid of their RSS readers all together as well. Even though we’re just getting started, I think we may be on to something.

Ok, Finally Finished

So that’s a lot of freakin’ words over three big ass posts talking about why laterstars is what it is. I’m glad to have finally gotten all my thoughts out there about this stuff, and I hope it gives you a better idea of what went into the decisions that were made, and some insight into why things you think should be different are how they are. Maybe it’ll even give you a reason to give those things a chance for a while. Of course, I know it’s not possible to be everything to everyone, so I’d love to hear from you about what you think, especially if I didn’t talk about something you’ve been wondering about.

Thanks for taking the time to read, and I look forward to checking out your links! Now back to your normally scheduled blog of feature announcements and reasonably sized posts…